Indian startup Emergent launches Wingman and jumps headfirst into the autonomous AI agents race
Emergent just shook up the AI agents market with a launch that gets straight to the point.
The Indian startup, headquartered in Bengaluru and known for democratizing vibe-coding — letting anyone build full applications without writing a single line of code — has introduced Wingman. It is an autonomous, messaging-based artificial intelligence agent that lives where most people already spend a big chunk of their workday: on WhatsApp, Telegram, and iMessage.
The play is smarter than it looks at first glance. Instead of building yet another new platform for users to learn and adopt, Emergent bet on messaging as the primary channel for interacting with the agent. The logic is simple: if work already happens through chat, why should an AI assistant be any different? When you think about it, it makes total sense — the learning curve practically vanishes when the interface is something you already use dozens of times a day.
Wingman is not here just to answer questions. It executes tasks, connects tools like email, calendars, and work software, manages routines, and even knows when it needs to ask for your approval before making a more important decision. This puts Emergent in a race that is heating up fast, alongside names like Anthropic, Microsoft, and the growing OpenClaw — previously known as Clawdbot and Moltbot.
As Mukund Jha, co-founder and CEO of Emergent, put it: the natural next step after helping people build software was helping them operate more autonomously through it. The idea is to move from software that merely supports the business to software that actively helps run it.
What Wingman actually does in your day-to-day
Wingman was designed to work as a real collaborator, not a glorified chatbot. The difference lies in autonomy with accountability — it can navigate systems, trigger APIs, organize workflows, and even coordinate complex task sequences, all within the messaging environment you already use. There is no separate dashboard to open, no new interface to learn, and definitely no elaborate technical setup required to get started. You simply send a message and the agent understands the context, interprets the intent, and moves to execution.
What makes this possible is the intelligence layer that Emergent built on top of language models. Wingman does not just process text — it reasons about what needs to be done, identifies dependencies between tasks, and organizes a logical sequence of actions to reach the expected outcome. If something along the way requires a decision that impacts sensitive data or involves an irreversible action, the agent pauses, notifies the user, and waits for confirmation before continuing.
Receive the best innovation content in your email.
All the news, tips, trends, and resources you're looking for, delivered to your inbox.
By subscribing to the newsletter, you agree to receive communications from Método Viral. We are committed to always protecting and respecting your privacy.
This mechanism is what Emergent calls trust boundaries — confidence limits that define how far the agent can act on its own and at what point it needs human approval. It is one of the most important differentiators of the product because it puts control back in the hands of the person using it, without compromising workflow fluidity. In a world where a lot of people are still wary of handing full autonomy to AI systems, this kind of granular approach helps build trust progressively.
In practice, this means you can ask Wingman to reschedule meetings, consolidate information from different sources, generate reports, fire off notifications to teams, monitor processes, and much more — all through a conversation on WhatsApp, Telegram, or iMessage. For teams that already operate in a distributed fashion and rely on messaging to coordinate work, this agent fits naturally into the existing flow without creating friction or demanding a change in habits. It is exactly this kind of zero-effort adoption that separates products that become part of the routine from those that get forgotten after a week of testing. 🚀
Vibe-coding as the foundation for all of this
To understand why Wingman matters, it helps to take a step back and remember what Emergent built before it. The company gained attention by leading an approach that became known as vibe-coding — a way of developing software where the user describes in natural language what they want to build, and the platform turns that description into a functional application. No syntax, no development environment, no tech stack to configure. You describe it, the AI builds it.
This platform competes directly with tools like Cursor and Replit, but differentiates itself by focusing on people without a technical background who want to create complete full-stack applications from natural language prompts. This opened the doors of software development to a much larger audience than the traditional one, democratizing digital creation in a way that previous low-code tools were never able to match in depth.
The numbers behind Emergent’s vibe-coding platform are impressive: more than 8 million builders have already used the tool to create and publish software, with over 1.5 million monthly active users. Founded in 2025, the startup raised 70 million dollars in January of this year, reaching a valuation of 300 million dollars, with heavyweight investors like SoftBank, Khosla Ventures, and Lightspeed Venture Partners at the table. These are numbers that give the company serious runway to expand into new fronts — like this one in autonomous agents.
That DNA of radical simplification is baked directly into Wingman. The philosophy is the same: strip the technical complexity out of the way and put the power in the hands of whoever has the need, not necessarily whoever has the technical know-how. If vibe-coding eliminated the barrier between the idea and the software, Wingman is eliminating the barrier between intent and task execution within an AI agents environment. They are two different products that share the same worldview about how technology should work for people — and not the other way around.
The most interesting part is that these two Emergent products complement each other in a very direct way. Applications built through vibe-coding can be connected to Wingman as tools the agent triggers. This creates an ecosystem where the user can build their own custom solutions and then automate the operation of those solutions through the messaging agent. It is a value proposition that goes beyond the standalone product — it is a complete platform for anyone who wants to put AI agents at the center of their workflow without depending on an engineering team for every step of the way. 💡
The AI agents race is just getting started
The launch of Wingman comes at a time when the AI agents market is in full swing. According to industry analysts, we are entering the era where clicking buttons is being left behind — that is what Bret Taylor from Sierra says, one of the most respected names in the industry. The core idea is that interaction with software is migrating from traditional graphical interfaces to conversations with intelligent agents that execute actions on behalf of the user.
Anthropic has been betting heavily in this direction, with tools like Cowork, which offers Claude Code capabilities without requiring programming knowledge, and with agents that can navigate graphical interfaces the way a human would. Microsoft keeps expanding its Copilots across practically every product in the Office and Azure ecosystem, integrating agents directly into the tools that businesses already use every day — and is already working on its own OpenClaw-style system. And newer players like OpenClaw are emerging with specific approaches and gaining traction among early adopters, with a more experimental and playful philosophy for building agents.
The differentiator that Emergent brings to this fight is precisely the channel. While most competitors are building experiences within their own platforms or inside specific enterprise software, Wingman goes to where the user already is. Messaging as an interface is not a new idea — WhatsApp-based assistants have been around for years — but doing it with the depth of autonomy and reasoning that current language models enable is something qualitatively different from what existed before.
Mukund Jha explained the decision to TechCrunch pretty directly: a lot of real work already happens through chat, voice, and email — requesting something, following up, sharing context, making a decision. And increasingly, these will be the primary channels through which people interact with AI agents as well. The speed at which Large Language Models have evolved over the past two years has transformed what is possible inside a chat conversation, and Emergent is capitalizing on exactly that moment.
The limitations Emergent openly acknowledges
One point worth highlighting is Emergent’s transparency about Wingman’s current limitations. The CEO himself admitted that the system still struggles with highly ambiguous situations, confusing edge cases, unclear objectives, or workflows that depend heavily on human judgment. This honesty is refreshing in a market where many companies sell their agents as near-magical solutions without acknowledging the gaps.
The reality is that no current AI agent is perfect. Language models, no matter how advanced, can still misinterpret ambiguous instructions, lose context in long conversations, or make suboptimal decisions when the situation goes off-script. Wingman’s trust boundaries help mitigate the risk of these failures, but they do not completely eliminate the need for human oversight — especially in critical scenarios. Knowing this before adopting the tool is essential for setting the right expectations and using the agent the way it actually works well.
This acknowledgment of limitations also reinforces the governance approach that Emergent chose. Instead of promising an agent that does everything on its own, the company opted for a progressive trust model where the user gradually expands the scope of Wingman’s autonomy as they gain confidence in the system. It is a more mature and more sustainable approach in the long run. 🧠
Access model and availability
Wingman is launching with a limited free trial period, after which access will be paid. Users who already have an account on Emergent’s vibe-coding platform can use the agent directly through their existing accounts, which makes the transition easier and lowers the barrier to entry for those already in the company’s ecosystem.
Full pricing details have not been released yet, but the strategy of offering a trial before charging is smart for a product in this category. Autonomous agents need to demonstrate real value before a user commits financially, and a trial period lets each person evaluate whether Wingman truly fits into their workflow.
What to expect from the next steps
Emergent has not yet shared a detailed public roadmap for Wingman, but the company’s track record suggests new integrations and capabilities will arrive at a rapid pace. The nature of modern AI agents allows new tools to be connected relatively quickly, which means the range of actions Wingman can execute is likely to grow as the user base expands and real-world use cases emerge. Feedback from people using it day-to-day is direct fuel for development with this type of product, and Emergent has a history of iterating fast based on what it learns from its millions of users.
From a technical standpoint, the most interesting challenge the company will face is maintaining reliability and security as Wingman gains more autonomy and accesses more critical systems. An agent that schedules meetings carries a very different risk level than an agent that can execute transactions, modify production data, or make financial decisions. Balancing autonomy with governance is one of the toughest frontiers in AI agents development, and how Emergent navigates this will be one of the most important indicators for evaluating the product’s maturity over time.
The 70 million dollar investment and backing from investors like SoftBank and Khosla Ventures give the company an important financial cushion to keep investing in research, infrastructure, and expansion. In today’s AI market, having the capital to iterate fast and scale safely makes all the difference between being a relevant player and falling behind in the race.
It is still too early to say whether Wingman will establish itself as a benchmark in the segment, but Emergent’s bet has clear strategic consistency. The company is not trying to compete with Anthropic or Microsoft on the same turf — it is creating a different playing field, where adoption is easier, the learning curve is nearly zero, and value shows up fast. For startups, freelancers, small teams, and professionals who live on WhatsApp during work, an AI agent that arrives through the same channel where conversations already happen has a much more direct proposition than any fancy dashboard.
What is already clear is that the combination of vibe-coding, messaging, and autonomous agents that Emergent is assembling points to a vision of the future where anyone, regardless of technical background, can have a truly functional AI assistant working for them. Not as a distant future promise, but as something available today, in the messaging app that is already open on your screen. And in the tech market, simplicity that works tends to beat complexity that impresses. 🤖
